Prospective studies place the frequency of major stroke following surgery utilizing cardiopulmonary bypass (CPB) at 5%, with half the patients so affected suffering long-term disability. Neuropsychologic performance is acutely impaired in 20-60% of cardiac surgical patients, with 7-15% of patients permanently affected. Our long-term objective is to understand the pathogenesis of these injuries, and to develop techniques and therapies to eliminate or ameliorate them. We believe the process of CPB is likely to affect outcome of neurologic insults which occur during CPB. This process affects the cerebral vasculature, pumping functions, and the blood itself. Hemodilution, hypothermia, non-pulsatile flow, and blood gas management may each compromise compensatory mechanisms which ordinarily maintain neuronal viability, namely increases in cerebral blood flow (CBF) and oxygen extraction. Our specific aims are to determine whether the following affect cerebral oxygen availability and autoregulatory processes during CPB: 1) hypothermia (25 degrees C) alone; 2) hypothermic blood gis management (alpha-stat vs pH-stat); 3) blood rheology (pentoxitylline); and 4) pulsatile perfusion. Using information derived from these studies, specific bypass management strategies will then be tested to attempt to ameliorate acute neurologic injuries. These questions will be addressed using a rabbit model of CPB we have developed. The cerebral circulation will be "challenged" with either progressive hypoxemia or hypotension during CPB. Oxygen availability will be assessed by measurements of cerebral oxygen extraction, blood flow, metabolic rates for oxygen and glucose, plus electrophysiology. CBF is measured via individually validated microsphere technique, and cerebral metabolism as the product of CBF and cerebral arteriovenous content difference. We anticipate pH-stat management, pentoxifylline, and pulsatile flow will improve oxygen availability during hypothermic CPB. We will then use a transorbital middle cerebral artery occlusion to produce focal cerebral ischemia and attempt to use knowledge thus far gained to find optimal CPB technique with respect to neurologic outcome. The latter will be assessed using standard techniques to quantify infarct volume. We believe understanding of the cerebral physiology of CPB will permit development of techniques which ameliorate neurologic injury.